ST has integrated six accelerometers into a 2.5 x 3mm inertial measurement unit for sports wearables. “Capable of measuring events ranging from slight movements to intense impacts with equal accuracy,” said ST. The “low-g sensor provides high accuracy with limited range [and the] high-g sensor offers an extended range with less accuracy”. Packaged in a 0.86mm thick LGA and called ...
STMicroelectronics is aiming at healthcare wearables with an IC that senses bio-potentials and acceleration, and includes an AI processor and a finite state-machine for pedometer, step detection and step counting. “The machine-learning core and finite state machine enable product designers to implement simple decision trees for neural processing on the chip,” according to the company. “ST1VAFE3BX provides opportunities to extend ...
Raspberry Pi 5 now has the option of an AI co-processor for neural networks, artificial intelligence and machine learning. The AI kit comprises the same M.2 HAT+ recently announced for holding NVMe memory drives, instead loaded with 13Top/s Hailo-8L AI processor. Connection, according to Raspberry Pi, is over its single-lane “PCIe 3.0 connection running at 8Gbit/s” – which is the ...
Arduino has bought out second revisions of its Nano 33 BLE and Nano 33 BLE Sense bluetooth-equipped microcontroller boards, keeping the sensor mix, but changing several of the sensor ICs. The nRF52840 IC (via a u-blox NINA-B306 module) stays the same, and still provides both an Arm Cortex-M4 processor and a Bluetooth transceiver – to go with these is an ...
Rohm has developed a 30mW artificial intelligence IC with on-device learning capability for edge processing – 30mW here is an example power, dependent on algorithm running. “Until now it has been difficult to develop AI chips that can learn in the field consuming low power,” according to the company. This one is intended to use “artificial intelligence to predict failures ...
Aaeon has introduced its first autonomous robot controller, with integrated hardware time synchronisation and support for an external nine-axis IMU (inertial measurement unit). Aimed at outdoor autonomous mobile robots and called RBX-I2000, “the largest improvements offered are its speed and positional accuracy,” according to the company. “By moving away from solely software-based time synchronisation, latency is reduced from the standard ...
Imagination Technologies is offering early-stage companies access to IP for four of its PowerVR Series8XE GPUs and three Series3NX neural network accelerators without any licensing costs. “This lowers the barriers of entry to SoC design, enabling scale-ups to create IoT and AI products for applications covering smart home, industrial designs for smart cities or smart factories, smart kiosks and signage ...
An edge inference accelerator developed by Flexlogix has a 4k MAC dynamic tensor processor array and is optimised for Mpixel image processing models in medical, surveillance and IoT applications. The InferX X1 edge inference accelerator is designed for processing real-time Mpixel vision workloads which requires high bandwidth support for deep learning models which operate with small batch sizes in real ...
Accelerator on a Chip International Program aims to create an electron accelerator on a silicon chi
Xsens has released an upgraded 12 x 12mm versions of its MTi 1-series of motion-sensing inertial measurement unit (IMU) modules. Roll, pitch and yaw measurement accuracy has been improved, according to the firm, and tolerance to torsional mechanical stress has been increased compared with first-generation products for demanding applications subject to shock or vibration. Static roll/pitch measurement accuracy is now ...
Electronics Weekly
